commit a6fa941d94b411bbd2b6421ffbde6db3c93e65ab upstream.

Don't mess with file refcounts (or keep a reference to file, for
that matter) in perf_event.  Use explicit refcount of its own
instead.  Deal with the race between the final reference to event
going away and new children getting created for it by use of
atomic_long_inc_not_zero() in inherit_event(); just have the
latter free what it had allocated and return NULL, that works
out just fine (children of siblings of something doomed are
created as singletons, same as if the child of leader had been
created and immediately killed).

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20120820135925.GG23464@ZenIV.linux.org.uk
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

An rmdir pushes css's ref count to zero.  However, if the associated
directory is open at the time, the dentry ref count is non-zero.  If
the fd for this directory is then passed into perf_event_open, it
does a css_get().  This bounces the ref count back up from zero.  This
is a problem by itself.  But what makes it turn into a crash is the
fact that we end up doing an extra dput, since we perform a dput
when css_put sees the ref count go down to zero.

css_tryget() does not fall into that trap. So, we use that instead.

Reproduction test-case for the bug:

 #include <unistd.h>
 #include <sys/types.h>
 #include <sys/stat.h>
 #include <fcntl.h>
 #include <linux/unistd.h>
 #include <linux/perf_event.h>
 #include <string.h>
 #include <errno.h>
 #include <stdio.h>

 #define PERF_FLAG_PID_CGROUP    (1U << 2)

 int perf_event_open(struct perf_event_attr *hw_event_uptr,
                     pid_t pid, int cpu, int group_fd, unsigned long flags) {
         return syscall(__NR_perf_event_open,hw_event_uptr, pid, cpu,
                 group_fd, flags);
 }

 /*
  * Directly poke at the perf_event bug, since it's proving hard to repro
  * depending on where in the kernel tree.  what moved?
  */
 int main(int argc, char **argv)
 {
        int fd;
        struct perf_event_attr attr;
        memset(&attr, 0, sizeof(attr));
        attr.exclude_kernel = 1;
        attr.size = sizeof(attr);
        mkdir("/dev/cgroup/perf_event/blah", 0777);
        fd = open("/dev/cgroup/perf_event/blah", O_RDONLY);
        perror("open");
        rmdir("/dev/cgroup/perf_event/blah");
        sleep(2);
        perf_event_open(&attr, fd, 0, -1,  PERF_FLAG_PID_CGROUP);
        perror("perf_event_open");
        close(fd);
        return 0;
 }

Signed-off-by: Salman Qazi <sqazi@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Tejun Heo <tj@kernel.org>
Link: http://lkml.kernel.org/r/20120614223108.1025.2503.stgit@dungbeetle.mtv.corp.google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>

The @func callback was invoked twice for group leader when
perf_event_for_each() called. It seems the commit 75f937f24bd9
("perf_counter: Fix ctx->mutex vs counter ->mutex inversion") made the
mistake during the change.

Signed-off-by: Namhyung Kim <namhyung.kim@lge.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1338443506-25009-1-git-send-email-namhyung.kim@lge.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

This reverts commit cb04ff9ac424 ("sched, perf: Use a single
callback into the scheduler").

Before this change was introduced, the process switch worked
like this (wrt. to perf event schedule):

     schedule (prev, next)
       - schedule out all perf events for prev
       - switch to next
       - schedule in all perf events for current (next)

After the commit, the process switch looks like:

     schedule (prev, next)
       - schedule out all perf events for prev
       - schedule in all perf events for (next)
       - switch to next

The problem is, that after we schedule perf events in, the pmu
is enabled and we can receive events even before we make the
switch to next - so "current" still being prev process (event
SAMPLE data are filled based on the value of the "current"
process).

Thats exactly what we see for test__PERF_RECORD test. We receive
SAMPLES with PID of the process that our tracee is scheduled
from.

Discussed with Peter Zijlstra:

 > Bah!, yeah I guess reverting is the right thing for now. Sad
 > though.
 >
 > So by having the two hooks we have a black-spot between them
 > where we receive no events at all, this black-spot covers the
 > hand-over of current and we thus don't receive the 'wrong'
 > events.
 >
 > I rather liked we could do away with both that black-spot and
 > clean up the code a little, but apparently people rely on it.

Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: acme@redhat.com
Cc: paulus@samba.org
Cc: cjashfor@linux.vnet.ibm.com
Cc: fweisbec@gmail.com
Cc: eranian@google.com
Link: http://lkml.kernel.org/r/20120523111302.GC1638@m.brq.redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>

Merge reason: We are going to queue up a dependent patch:

"perf tools: Move parse event automated tests to separated object"

That depends on:

commit e7c72d8
perf tools: Add 'G' and 'H' modifiers to event parsing

Conflicts:
	tools/perf/builtin-stat.c

Conflicted with the recent 'perf_target' patches when checking the
result of perf_evsel open routines to see if a retry is needed to cope
with older kernels where the exclude guest/host perf_event_attr bits
were not used.

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

We can easily use a single callback for both sched-in and sched-out. This
reduces the code footprint in the scheduler path as well as removes
the PMU black spot otherwise present between the out and in callback.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/n/tip-o56ajxp1edwqg6x9d31wb805@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>

We always need to pass the last sample period to
perf_sample_data_init(), otherwise the event distribution will be
wrong. Thus, modifiyng the function interface with the required period
as argument. So basically a pattern like this:

        perf_sample_data_init(&data, ~0ULL);
        data.period = event->hw.last_period;

will now be like that:

        perf_sample_data_init(&data, ~0ULL, event->hw.last_period);

Avoids unininitialized data.period and simplifies code.

Signed-off-by: Robert Richter <robert.richter@amd.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1333390758-10893-3-git-send-email-robert.richter@amd.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>

No need to have an additional function layer.

Signed-off-by: Robert Richter <robert.richter@amd.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1333643084-26776-4-git-send-email-robert.richter@amd.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>

In perf_event_for_each() we call a function on an event, and then
iterate over the siblings of the event.

However we don't call the function on the siblings, we call it
repeatedly on the original event - it seems "obvious" that we should
be calling it with sibling as the argument.

It looks like this broke in commit 75f937f24bd9 ("Fix ctx->mutex
vs counter->mutex inversion").

The only effect of the bug is that the PERF_IOC_FLAG_GROUP parameter
to the ioctls doesn't work.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1334109253-31329-1-git-send-email-michael@ellerman.id.au
Signed-off-by: Ingo Molnar <mingo@kernel.org>

Merge reason: we need to fix a non-trivial merge conflict.

Signed-off-by: Ingo Molnar <mingo@elte.hu>